Scribd
Upload
3 views

Stat Extra Examples

Exercise of statistics

Uploaded by

eshaansariansari452
Copyright
© © All Rights Reserved
Available Formats
Download as PDF or read online on Scribd
3 views

Stat Extra Examples

Exercise of statistics

Uploaded by

eshaansariansari452
Copyright
© © All Rights Reserved
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 9
5. S. Shah & Sons | EXERCISE 2.5] Section F - 5 Marks rind correlation coefficient by Karl p ‘earson's fi ormula and and interpret it umber of 80 | 95 | 105} 130} 135 | 137} 140 | 4g 0) 176 D) imo) 1% 70} 75} 100 102) 110 | 125 el _— 145 [Ans. : r= 0.97, Significant positive correlation ] Find product moment correlation coefficient : ” —| 62) 67 67 fy 2 oT | {I Salein 62 162 | 62] 6 Sy (1000 units) : nt Fespentione Fag tye] asf vo} so} aa] ay ST igh | I 10 Rs) Fin 100 | [Ans, tp 1 Ny ; : sHlicient ; i Find Pearson's correlation coefficien | x= Price of water (Rs.) | 20 | 46 [4 [32 Fag al nlp y= Demand (Thousand Packets) | | 42 | 54 XD | 0} 58} 46 % | | [Ans, : ; 0% | Section D - 3 Marks | 8. Find correlation coefficient from the following data ; | W810; Sx 120; Y y= 250; Sex —15)2 130; 3 (y~25)? <99, | £(x=15)(y-25) = —49 [An - 067 | 9. Find correlation Coefficient from the following data : | melds Sx “198; Sy= 240; ¥(x 15) =315; S(y—21)2 ~345, XU(x=15)(y~21) = 276 [Ans. : r= 0.95} 10. Find correlation Coefficient, n=10;¥= 21,9 = 19, X(e= 20? =160; 3(y~20)? = 276; X(x=20)(y~20) = go [Ans. - 05] Section F - 5 Marks Mustration 14: Find Correlation coefficient from the following data: © Annual income : Per capita (Rs. 100) | 300] 350 400] 450] y= Price Index 1.24] 1.27] 1.18] 1.20] 1. } 300 | 350 400 450 | 500 550 «0 650 700 4500 a yn =500 p=N2 a2 Take u= —- and v= ee =100y—122 pe Edy Vn? (0)? xfnEv ey? {9x 60-(0)? xf9x90-(4)? 36 1540 x 794 =-—_36_ 23.24 x 28.18 36 * 654.90 = 0.05 Illustration 15 : Find correlation coefficient between production of ‘ars and their prices from the following data : Production of Cars (in 1000 units) 3.5| 4.2] 5.6| 6.5 | 7.0] 8.2] 8.8) 9.0 | 97 10.0 8.1 .2| 8.0 | 8.0 (in 1,000 Rs.) 9.8 | 9.0 s| 84 8.3| 8.2] 8.2 : iy son of ears by x and the Price Of a car py »7 as, 2710 ( _ 82 = 4727 |292 | _ +5 25 = 1077 n>Duv-Dudv 10(—1077)-(25)(—2) - 1 10(4727)- (25) xy 10(292)—(-2)? -10770+50 ~ 47270 625 x [2920-4 4 ___-10720 46645 x 2916 _ _=10720 215.97 x54 =-0.92 Meaning : There is a high di i Jation be : . egree negative correlal Production of cars and Price of a car. ° ~ ~ B.S. Shah & Son: [EXERCISE 2.6 | mens Soe Section F . 5 Marks | Find Karl Pearson's Coefficient of Correlation « | <= Use of fertilizer (quintaly — [5 112520 Taey | y= yield of wheat per hectare | *(quintaly 46 | 56] 49 salar | 7 reat Ans. : r = 999) | Find correlation Coefficient from the following is fee | Advt. cost 7 T ——___ | | | 2800| 2000} 2100| 2200 za 5 2400 | 2300 | 190) 2109 | 28 | 18 | 18 | 19 20 | 24 | 20 | 18} 17] 18 [An data : 9} 8.4 | 1.4 7.71 5.6] 9.11105] 50} 125] 25] 200] 100) 175) 225 +r = 0.92] | Find correlation coefficient from the following x = Monthly income (Rs.10,000) y = Monthly cost (Rs. 100) [Ans. : r = 0.91] | Find Karl Pearson's coefficient of correlation « = Sales of fans ] (Thousand units) | 25}29}3.1| 26 | 41 | 46 3.7) 39 = Profit (Thousand Rs.) 450 | 500 | 600 | 550 | 700 | 650 | 550| 650| so0| 400 3.4| 36 Section C - 2 Marks perties of correlation coefficient : |. Thecorrelation coefficient always lies between—land+li.e.-I srt. 2. The correlation coefficient does not depend upon aul ie. If heights and weights are expressed in cm and kg aa sera Coefficient is found out and if they are expressed in inches an 7 ate the correlation coefficient is found out, the value is the same in bo! . t 3. The correlation coefficient between x and y Is the same a: Correlation coefficient between y and x ie. r(x, y) =r (y, xh is the ater Charenton ALT exnmelations coritioient fs Extependient of chang, THe ooerelationt — OF ong: A ’ Ry & sen Le Ie ee variadle vis changed (a variable y, by way = te > QO). then the core, ff aan» IS Changed to v Dy ti Nethiy, ben Posing a>Qyeond) / & We the sas of all the Values of one Variable are changed, the - ofte om ation coefficient is changed. ie. (=x, y) = ny =) == ny) & Thewh sent Depween x aed is The same as the correlation Re S ‘and Kia raver Qn), Her, a and ¢ must Wwe AWS \ "a, y) s benveen O and I Le. Ose? s1. of correlat coefficient ris the same as the sign of Corny) Section C - 2 Marks lnterpretation of correlation coefficient 7: On coefficient is based on the ASSuMpLiog m of Ovo variables v and y either lie on one ome line, Hence, the correlation Coetfficien: found out after verifying that the information ows More or less linear relationship, taneous chat in the values of two variables and if and effect relationship between the two. variables, then aship is said to be correlation. Hence, before studying Sorrelaton between nwo variables it sho ould be ascertained that the relationship between s linear and there exists cause and effect relationship between the variabl The relationship between the number of vehicles and Detween income and expenditure, and that between ‘ue tO causation, and hence useful results can be Production of cars in Japan and the figures of population in India are studied together they will show: simultaneous changes in the same direction, From this, it cannot be said that there is correlation between the production of cars in Japan and population: of India, Ss there is no cause and effect relationship between tem. This pe o comelation is meaningless or ts. that rainfall and yield are dh Shah & Sons _ 41, the linear correlation between the variables is perfect when / ariabl » ani ‘or such variables the points of the scatter diagram are in on? Fe re. . , rm joing upward from lower left corner to the upper right commer. * 1c i va hese points is of the form y = a + br, (b > 0). When two es have perfect positive linear relationship, an increase in the value one variable is associated with a proportional increase in the value of note variable. retation of r=-l: 1, the linear correlation between the variables is perfect regative- For such variables the points of the scatter diagram are in one fersight HiME going down word from upper left comer to the lower right comer. ‘The line joining these points is of the form y = a — bx, (b > 0). When two vyariables have perfect negative, linear relationship, an increase in the value of one variable is associated with a proportional decrease in the value of another variable. joterP when, 7 = ~ interpretation of r= 07 0, there is no linear correlati ‘e randomly distribute ion between the variables. For such .d in scatter diagram. When the value nt r = 0, it can only be said that there is no he variables. There can exist any other type of point should be kept in mind while When r= variables the points ar f the correlation coefficie linear relationship between t lationship between these variables. This interpreting r = 0. interpretation of partial correlation coefficient + Ifthe correlation coefficient is near to 1, it can be said that there is hi legree of positive correlation between the variables. But if the value ofr sitive and near to 0 it can be said that there is insignificant pos ‘orrelation between the variables. Similarly. if the value of r is near to — indicates high degree of negative correlation between the variables ar regative and near to 0 the correlation between the variable is ins gative. In reality the correlation coefficient near to zero indi ‘ence of correlation between the variables. ) Spearman’s method of rank correlation : ae Charles Edward Spearman has given another methos © rei ee coefficient between two variables in 1904. In this method nd hence values the ranks are used to find out correlation coefficion , the method is known as the method of rank correlation. Wo nding A or ree lt jation _—— numerically expressed, 5, ~ : - alitative phenomena cannot be » there are 10 com * but know that quar them ranks: €-B suppose . "Petitors nt to assign the snient {0 give marks to these competitors, Ins, nye ad ranks as first rank, second rap) : judges have given ranks to the Linear Correl convene if beauty contest. Iris of that they can be eas! according to their merits. then we may K et . , be interested in knowing how far the two ia ‘This can be measured by coefficient of ra agree in assi setics like sty, intellig eh tion, Qualitative characteristics like honesty, intelligence, Poverty correlati i but they can be easily assigned ranks accor fi re: bers, difficult to express in num ; in ‘ ‘i ita ye found out betw 7 to their merit and correlation coefficient can een ranks Spearman's rank correlation method. also useful when there are more variations among tf method becomes tedious in sy, participants, gning ranks. This method is observations of any series. Pearson’s situations. For finding correlation coefficient of such data, the ranks are give, separately to the values of each series. In each series the highest value is first rank. The next value is given the second rank and so on. If two values a equal then there isa tie. In such a case both values are bracketed and they = given ranks equal to the average value of their ranks. e.g., suppose the marks of 8 students in statistics are 63, 65, 87, 24, 61, 57, 50, 61. The student getti 87 marks will be given first rank. The second rank will be given to the & vi getting 65 marks, third rank will be given to the student getting 63 marks. i - there are two students getting 61 marks. i.e. there is a tie. Both these ude can be given average rank of ranks 4 and 5 =4*5— setin 57 . " ‘ 7 74-5 each. The next student ig 57 marks will be given sixth rank and so on. After assigning ranks in two seri i series, the di ranks of the two series are obtained for all tt denoted by d and the value of Ya? correlation is then calculated by the folk ifferences in the corresponding he pairs. These differences are is obtained. The coefficient of rank lowing formula : 6ya* ~ n(n? 1) Where, n= Number of pairs, Tie in observatio ns [when ob: i observations of x servations are equal] : f (some observatio: a a a We say that it is a ne aa i tie a correlation coefficient. This Correction is necessary in the a la of rank EMS Correction is to be done by Corrector Factor (Ch) ion } B. S. Shah & Sons for tied ranks (when some observations are equal), the So, ng rank correlation can be written as follows : ning 6 Sa? + cF L - goes (Fai) ( nm) where, CF = Correction Factor = | formula for ) and m = Number of times observation is repeated. The value of correlation coefficient obtained by this method also lies een -1 and +1. If the ranks are same for each pair of two series, then each of dis 0. Hence, yd? =0 and the value of r = +1, which shows perfect itive linear correlation between the two variables. If the ranks are exactly reverse order for each pair of two series, the value of r = —1, which shows fect negative linear correlation between the variables. Section D - 3 Marks NN ee ee a a a

You might also like